Short communication: Meta-analysis of dairy cows fed conventional sorghum or corn silages compared with brown midrib sorghum silage.

A meta-analysis was conducted to compare the effects of feeding dairy cows conventional sorghum silage (CSS) or conventional corn silage (CCS) compared with brown midrib sorghum silage (BMRSS) diets on dry matter intake (DMI), milk production, and milk composition. Data from 9 published articles (1984 to 2015) were used to contrast diets with CSS (7 means comparisons; 104 cows) or CCS (13 means comparisons; 204 cows) versus BMRSS diets. Statistical analysis was performed using fixed or random effects models with the Metafor package of R (https://www.R-project.org). The degree of heterogeneity was measured with the I2 statistic, and publication bias was determined with funnel plots and Egger's regression test. Other sources of heterogeneity of response were analyzed through meta-regression. Estimated effect size was calculated for DMI, milk production, and milk composition. No evidence of publication bias was observed for any variable tested. The highest degree of heterogeneity (I2 = 41.5 and 72.6%) was observed for DMI among dependent variables tested in both comparisons, indicating that intake responses to silage type are rather inconsistent; in contrast, milk production had the lowest degree of heterogeneity (I2 = 0%), supporting the idea that the responses of this variable to silage type were very consistent across studies. Compared with BMRSS diets, cows fed CSS diets exhibited decreased milk production (1.64 kg/d), milk fat concentration (0.09%), milk fat yield (0.08 kg/d), milk protein yield (0.04 kg/d), and milk lactose yield (0.16 kg/d) and tended to decrease DMI (0.83 kg/d). Compared with CCS diets, cows fed BMRSS diets increased milk fat concentration (0.10%), but decreased milk protein concentration (0.06%) and tended to increase lactose yield (0.08 kg/d). Meta-regression indicated that days in milk affected DMI and milk production when CSS diets were compared with BMRSS diets, and DMI when CCS diets were compared with BMRSS diets. Additionally, the inclusion rate of silage in the diet and dietary neutral detergent fiber affected yields of milk fat and lactose, respectively, when CCS and BMRSS diets were compared. Overall, lactation performance improved when cows were fed diets formulated with BMRSS compared with CSS, but performance was not different for cows fed BMRSS and CCS diets. However, the small sample size may have influenced these results by increasing the margin of the error and, concurrently, the power of the meta-analysis. Results of this analysis suggest that additional research is needed to explore the effects of days in milk and the inclusion rates of silages in the diets when comparing BMRSS with CSS or CCS.